Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Plan-then-execute implementation against SPEC.md. Native single-thread loop, no sub-agents. On test or build failure, auto-invokes the backprop skill before retrying — a failed verification always considers whether a new §V invariant would prevent recurrence. Triggers when the user asks to build, implement, execute the spec, or tackle a specific §T task (`build §T.3`, `build --next`, `implement next task`, `run the build`). Expects SPEC.md to exist; if not, defers to the spec skill.
| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -70% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 38% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -11% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -11% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -42% | 0% |
Single-thread native plan→execute. You are main Claude. No swarm.
SPEC.md. If missing → tell user to invoke the spec skill first. Stop.FORMAT.md once if not loaded.§T.n → that task only--next → lowest-numbered row with status . or ~--all or empty → every . row in §T orderHigh blast radius (shared module, auth, data, money, public §I)? Run /review first. Trivial & reversible? Skip planning ceremony, just do step EXECUTE.
Native plan mode — you delegate to it, you do not reinvent task breakdown. For chosen task(s):
prove each §V touched. Which test, not "add tests". "Do TDD" alone backfires; the spec says what to check. Each §V touched → a named test that fails first.
Show plan. Wait for user OK unless auto mode.
Per task in order:
. → ~. Just write to SPEC.md.~ → x. Next task.On test/build failure:
bug: <cause> first, let it update §V and §B, then resume build against updated spec.Rule: never silently fix root-cause without considering backprop. §B is the memory that stops recurrence.
T<n>: <goal line> + §V cites.Task x only if:
cat SPEC.md | grep §T is the dashboard.Other measured skills in the registry, with their headline benchmark lift.